Personalized medicine is on the way (but not quick enough)

The New York Times had a great article a couple of days ago on the need for personalized medicine to become more than a catchy phrase.

As we're learning more about the interaction between genes and drug metabolism, we're also learning that large numbers of people are either taking the wrong drug or taking drugs that won't work.

tags: ,

Researchers have known for some time that genetic variants determine how well drugs work. Some versions of a gene cause a drug to be metabolized faster, some slower, and when combined, at an intermediate rate. In the simplest case, the slow metabolizers might be getting too much of a drug, and the fast metabolizers too little. Sometimes the drug won't work in people with the wrong genotype, and sometimes the wrong drug is lethal.

Too little education?
Despite this knowledge, there's been a delay putting it into practice. One problem is education. The genetic tests are appearing quite rapidly, leaving physicians with little time to catch up.

According to Steve Murphy (The Gene Sherpa), who adds: "I see that lawsuit creeping closer and closer":

â¢Only 17% of patients with familial colon cancer risk were referred to appropriate genetic counseling according to a study at Harvard.

â¢Prenatal counseling is not offered to 9 of 10 women who meet indications!

â¢Only 37% of MDs know that BRCA genes can be passed by the father!

And from the NYT:

Moreover, doctors do not always conduct the tests or follow the results. The big insurer UnitedHealthcare found in 2005 that 8 percent of the women getting the drug had tested negative for the required genetic characteristic. An additional 4 percent had not been tested at all, or their test results could not be found.

Too little time?
There are also delays in translating information from the research lab to the clinical world. Researchers may know that a certain version of a gene will affect metabolism, but the implications for changing treatment aren't always clear. Clinical trials could answer those questions, but they are expensive and would likely increase drug costs.

Presently, there are only a few drugs with genotyping recommendations on their labels. Earlier this month, the FDA met to discuss modifying the label information for additional drugs. Some of the drugs under consideration are Erbutix and Vectibix, used to treat colon cancer. According to the NYT, these drugs are ineffective for 40 percent of patients with certain tumor genotypes.

Are patients losing patience?
Whatever happens, it seems that increasing the use of genetic information in drug prescribing can't come soon enough for many patients, especially women who've had breast cancer.

The NYT described a study at the Mayo Clinic where:

In the case of tamoxifen, Dr. Matthew P. Goetz of the Mayo Clinic and colleagues went back to an old trial and used stored tumor samples to test the 2D6 genes of each patient. The researchers reported in 2005 that 32 percent of the women with inactive 2D6 enzyme had relapsed or died within two years, in contrast to only 2 percent of the other women.

In other words, the dead women might have done okay, if they'd had a different drug.

From the NYT:

For more than two years, Jody Uslan had been taking the drug tamoxifen in hopes of preventing a recurrence of breast cancer. Then a new test suggested that because of her genetic makeup, the drug was not doing her any good.

"I was devastated," said Ms. Uslan, 52, who stopped taking tamoxifen and is now evaluating alternative treatments. "You find out you've been taking this medication for all of this time, and find out you are not getting any benefit."

Andrew Pollack, December 30, 2008, Patient's DNA May Be Signal to Tailor Medication, The New York Times.

More like this

Yes, it is true. Education is how this movement takes hold. But where's the money in that? Which is why we have seen some absolutely not ready for clinical use tests floundering. I blame the current medical education system, which does not require significant genetics or biostatistics prerequisites. That same system also fails to teach genetics. Why? Clinician sre NOT the teachers, therefore, students fall asleep. If they wanted hard core basic science, they would have done PhDs.

We need a systematic overhaul of the way we train doctors.


Personalized Cancer Medicine Is Here, Now

As we enter the era of "personalized" medicine, it is time to take a fresh look at how we evaluate treatments for cancer patients. More emphasis is needed matching treatment to the patient. Patients would certainly have a better chance of success had their cancer been chemo-sensitive rather than chemo-resistant, where it is more apparent that chemotherapy improves the survival of patients, and where identifying the most effective chemotherapy would be more likely to improve survival.

Findings presented at the Annual Meeting of the European Society for Clinical Investigation in Uppsala, Sweden and the Annual Meeting of the American Assoication for Cancer Research (AACR) in San Diego, CA concluded that "functional profiling" with cell-based assays is relevant for the study of both "conventional" and "targeted" anti-neoplastic drug agents (anti-tumor and anti-angiogenic activity) in primary cultures of "fresh" human tumors.

Cell-based Assays with "cell-death" endpoints can show disease-specific drug activity, are useful clinical and research tools for "conventional" and "targeted" drugs, and provide unique information complementary to that provided by "molecular" tests. There have been more than 25 peer-reviewed publications showing significant correlations between cell-death assay results and patient response and survival.

Many patients are treated not only with a "targeted" therapy drug like Tarceva, Avastin, or Iressa, but with a combination of chemotherapy drugs. Therefore, existing DNA or RNA sequences or expression of individual proteins often examine only one compenent of a much larger, interactive process. The oncologist might need to administer several chemotherapy drugs at varying doses because tumor cells express survival factors with a wide degree of individual cell variability.

There is a tactic of using biopsied cells to predict which cancer treatments will work best for the patient, by taking pieces of live "fresh" tumor tissue, applying different chemotherapy treatments to it, and examining the results to see which drug or combination of drugs does the best job killing the tumor cells. A cell-based assay test with "functional profiling," using a cell-death endpoint, can help see what treatments will not have the best opportunity of being successful (resistant) and identify drugs that have the best opportunity of being successful (sensitive).

Funtional profiling measures the response of the tumor cells to drug exposure. Following this exposure, they measure both cell metabolism and cell morphology. The integrated effect of the drugs on the whole cell, resulting in a cellular response to the drug, measuring the interaction of the entire genome. No matter which genes are being affected, functional profiling is measuring them through the surrogate of measuring if the cell is alive or dead.

For example, the epidermal growth factor receptor (EGFR) is a protein on the surface of a cell. EGFR-inhibiting drugs certainly do target specific genes, but even knowing what genes the drugs target doesn't tell you the whole story. Both Iressa and Tarceva target EGFR protein-tyrosine kinases. But all the EGFR mutation or amplificaton studies can tell us is whether or not the cells are potentially susceptible to this mechanism of attack. They don't tell you if Iressa is better or worse than Tarceva or other drugs which may target this. There are differences. The drugs have to get inside the cells in order to target anything. So, in different tumors, either Iressa or Tarceva might get in better or worse than the other. And the drugs may also be inactivated at different rates, also contributing to sensitivity versus resistance.

As an example of this testing, researchers have tested how well a pancreatic cancer patient can be treated successfully with a combination of drugs commonly used to fight lung, pancreatic, breast, and colorectal cancers. The pre-test can report prospectively to a physician specifically which chemotherapy agent would benefit a cancer patient. Drug sensitivity profiles differ significantly among cancer patients even when diagnosed with the same cancer.

The funtional profiling technique makes the statistically significant association between prospectively reported test results and patient survival. It can correlate test results that are obtained in the lab and reported to physicians prior to patient treatment, with significantly longer or shorter overall patient survival depending upon whether the drug was found to be effective or ineffective at killing the patient's tumor cells in the laboratory.

This could help solve the problem of knowing which patients can tolerate costly new treatments and their harmful side effects. These "smart" drugs are a really exciting element of cancer medicine, but do not work for everyone, and a pre-test to determine the efficacy of these drugs in a patient could be the first crucial step in personalizing treatment to the individual.

Literature Citation:

Weisenthal, L.M. Functional profiling with cell culture-based assays for kinase and anti-angiogenic agents Eur J Clin Invest 37 (suppl. 1):60, 2007

Nagourney, R.A. Functional Profiling of Human Tumors in Primary Culture: A Platform for Drug Discovery and Therapy Selection (AACR: Apr 2008-AB-1546)

By Gregory D. Pawelski (not verified) on 04 Jan 2009 #permalink

It is amazing that people are not diagnosed properly for what they should be treated with. I saw the other day that companion diagnostics market for cancer drugs is < $ 10 million, that is some 0.01% of the global cancer drug sales.

the availability and the interpretation of genetic informtation should be regulated and cautiously introduced to the general population as the potential for personalised medicine can be greatly damaged by opportunistic genetic tests available